We have shown that a variety of antimicrobial peptides (AMPs) and nuclear binding proteins that mimic chemokines also have the capacity to rapidly activate host immune responses. We have proposed calling these early warnings signals alarmins. Alarmins are characterized by having in vitro chemotactic or in vivo recruitment activity for cells expressing GiPCR, together with the capacity to interact with other receptors resulting in the activation of immature dendritic cells (iDC) into mature antigen-presenting capable of interacting with T lymphocytes. These alarmins, if administered together with an antigen result in considerable augmentation of both in vivo cellular and humoral immune responses to the antigen. We previously identified both alpha and beta types of defensins as alarmins with chemotactic and activating effects on immature dendritic cells (iDCs) and in vivo immunoadjuvant effects. Some of the beta defensins interact with the CCR6 chemokine receptor, others with CCR2, while alpha defensins interact with an as yet unknown G-Protein Coupled Receptors (GiPCR). Another antimicrobial peptide known as cathelicidin (LL37) and its murine homologue CRAMP are chemotactic for FPR2 receptors expressed on monocytes and precursors of iDC. Cathelicidins also induce the maturation iDC and are equally as potent adjuvants in vivo as alum. In addition, we have previously also identified eosinophil derived neurotoxin (EDN, a ribonuclease), granulysin from lymphocytes, lactoferrin from neutrophils and HMGB1, a nuclear binding protein, as functional alarmins. Although alarmins are structurally distinct, they are preformed and constitutively available. Alarmin are rapidly released from granules, cytosol or nucleus of leukocytes and epithelial cells or from injured cells. Alarmins can also be induced to be produced in response to proinflammatory stimulants by keratinocytes or epithelial cells lining the GI tract, GU tract and tracheobronchial tree. As such, alarmins probably represent an early warning system to alert the host defense to danger signals. During the past four years, we have also identified and characterized High Mobility Group Nuclesome-binding protein-1 (HMGN-1) as an extracellular alarmin that is a necessary mediator of lipopolysaccharide (LPS)-induced (TLR-4-dependent) immune responses. HMGN-1 has the capacity to recruit and induce the maturation of dendritic cells (DC) at sites of injection. HMGN-1 activates NF kappa B and multiple MAP kinases largely in a TLR4 dependent manner. Upon coadministration with antigens, HMGN-1 markedly enhances specific immune responses and has potent adjuvant effects favoring Th1 immune responses. Conversely, mice genetically engineered to be deficient in HMGN-1 had greatly reduced antigen specific immune responses even in response to antigens administered together with LPS. This immune deficiency of HMGN-1 knockout mice was associated with deficient recruitment of DC to sites of immunization and reduced cytokine production by DC. Thus, HMGN-1 which is largely derived from non-leukocytes (e.g. epithelial cells) plays a non-redundant critical role in the development of innate and adaptive immune responses. In a previous experimental study, we showed that intraperitoneal administration of anthrax vaccine with HMGN1 protein markedly increased the production of both primary and secondary protective IgG antibodies to anthrax toxin. During this past year we evaluated the properties of the antiphagocytic capsule consisting of polyglutamic acid, of Bacillus anthraces. This capsule is a major virulence factor and is an immunosuppressive inhibitor of dendritic cell maturation. We plan to couple HMGN-1 to this capsular antigen, because a mixture of the two proteins becomes an immunogenic stimulant and may potentially provide a better vaccine against anthrax. During the past year we also showed that HMGN-1 knockout mice exhibit reduced resistance to tumor (EG-7 or EL-4) challenge. Conversely,tumor cells (EG-7 or EL-4) when transfected to overexpress HMGN1 showed a marked reduction in the rate of growth in normal mice. These observations indicated that HMGN1 was capable of augmenting tumor immunity. Since HMGN1 is the most potent of the alarmins in inducing antigen specific immune responses, we have chosen to study it further as an antitumor adjuvant. To maximize the adjuvant effects of HMGN1, we covalently linked it to a gp100 melanoma tumor antigen. Antigens fused to adjuvants have been shown to be delivered more effectively to the appropriate intracellular compartments of antigen presenting cells (APC's) resulting in improved antigen processing and presentation and greater T cell activation. We have immunized mice with gp100 linked to HMGN1 in the form of plasmid DNA using gene gun technology. This succeeded in inducing about 70% of the immunized mice to be resistant to a challenge with B16 melanoma tumor cells. However, therapy of mice with this plasmid DNA, which had been injected with B16 melanoma tumor cells four days previously, failed to inhibit tumor growth. We therefore injected a recombinant HMGN1 protein directly intratumorally into CT26 colon tumors in mice to proximate the adjuvant and antigen. This therapeutic vaccine trial did have a significant beneficial effect in slowing the tumor growth and prolonging the survival of mice, but did not cure any of the mice. We have to improve the delivery of the tumor vaccine and employ it in conjunction with other antitumor therapies.